🧰The Full Agency Stack

First, how to read this

Every system below has three tiers: what most agencies start with, what works for a while, and what to run once you’re growing. The early tools aren’t bad. Google Docs is excellent software. It just isn’t an onboarding system, and the moment you treat it like one it starts costing you hours you don’t notice you’re losing.

Each early pick is a good tool doing a job it wasn’t built for. And every upgrade has a trigger — a specific number of clients, or people, or hours — where the old thing stops working. This isn’t a ranking. It’s a map of what breaks when.

Match the tier to where you actually are, not where you want to be. Paying for the top tier at two clients is how you end up with $400/month of software managing $4,000 of revenue.

The stack at a glance

Two of those come with footnotes. Contracts needs a second tool, and reporting has three good answers depending on your size. Both are explained below.

1. Client Onboarding

Why Google Docs breaks

You become the database. A doc per client means the answers live in fourteen different documents in fourteen different shapes, half of them half-finished. Nothing fires when they submit. Nothing tells you when they haven’t. You find out a week later, when you sit down to start the work and the brand assets aren’t there.

What Google Forms buys you — and where it stops

Structure, for free. Answers land in one spreadsheet, in the same shape, every time. That alone justifies the switch.

Where it stops: no payment collection, no scheduling, no signature. Branching works only off multiple-choice and dropdown questions, and it routes to whole sections rather than individual questions.

The one that actually bites: file uploads force the client to sign in to a Google account. On an onboarding form asking for logos and brand assets, that’s a drop-off point — and what you get instead is an email with attachments, which puts you right back in the inbox you were trying to escape.

The pick: Typeform

• Free — $0, 100 responses/mo, Typeform branding
• Basic — $29/mo ($25 annual)
• Plus — $59/mo ($50 annual), 1,000 responses, 3 seats

Setup detail that matters: build one form with branching by service type, not five separate forms. Five forms means five things to update every time your process changes, and they will drift.

2. Contracts

Why Google Docs breaks

Version drift. Someone duplicates the last client’s contract, forgets to change a clause, and now you have twelve agreements with twelve sets of payment terms and no reliable way to know which client got which.

And nothing is actually signed. “Reply to confirm” is not a contract you’d want to rely on.

What Canva buys you — and where it stops

It looks professional, which genuinely matters on a $10k proposal.

Where it stops: Canva has no e-signature capability at all. You are still emailing a PDF and asking someone to reply. Pro is also strictly single-seat — $180/year for one person, and adding a second person means the Business plan at $250/person/year.

The pick: Claude to draft, an e-sign tool to execute

Claude is only half the workflow. It drafts and reviews, but it cannot execute a signature. You need both.

Claude Pro — $20/mo ($17/mo billed annually).

What Claude is genuinely good at here isn’t writing a contract from a blank page. It’s reading the one you already use and telling you what’s missing:

Here’s the contract I currently send clients [paste]. I run a [type] agency doing [service] on [retainer/project] terms. Go through it section by section and tell me: what’s missing that agencies in my position usually regret leaving out, which clauses are ambiguous enough to be argued either way, and where I’m exposed on scope creep, late payment, and IP ownership. Rank the gaps by how much they could actually cost me.

Then e-sign. Cheapest genuinely usable option: Dropbox Sign — free for 3 signature requests/month, or Essentials at roughly $10/user/mo billed annually with unlimited requests.

3. Outreach & Lead Gen

Why LinkedIn alone breaks

It isn’t a system, it’s a habit. Volume depends on your mood that morning, nothing is tracked, and there’s no structured follow-up — which is where nearly all the replies actually come from.

Worth knowing before you spend: Sales Navigator ($99.99/mo) does not raise your connection request limit. It raises search results and InMail credits. Plenty of people buy it expecting more outreach and get more searching. Connection limits sit somewhere around 100–200 invites per week — LinkedIn has never published an official number, so treat that as community-observed rather than policy.

Why Mailchimp is the wrong middle tier here

This is the one thing on this page I’d most want you to read.

Mailchimp’s Acceptable Use Policy prohibits cold outreach. Their terms require you to point to an opt-in for every commercial email you send, and explicitly bar sending to “purchased, rented, third-party, co-reg, publicly available data, or partner lists of any kind.”

That phrase — publicly available data — covers scraped and enriched lists, not just bought ones. Import an Apollo export into Mailchimp and you’re in breach at the moment of upload, before you send a single email. Enforcement is account termination and list loss, not a warning.

So Mailchimp isn’t a weaker version of cold email. It’s the right tool for a newsletter to people who opted in, and the wrong tool for lead gen at any price.

The pick: Apollo

• Free — $0. 10,000 email credits on a corporate domain, but only 10 exports/month
• Basic — $59/mo ($49 annual), 1,000 export credits
• Professional — $99/mo ($79 annual), 2,000 export credits

The real constraint is export credits, not email credits. Ten exports a month won’t build one usable prospect list, so the free tier is for evaluating the product, not running on it. Organization tier carries a 3-seat minimum, which makes Professional the practical ceiling for a solo operator.

Setup detail that matters: send from a separate domain to your main one, and warm it before you scale. A burned primary domain takes your client email down with it. And check the rules where you and your prospects are — CAN-SPAM, GDPR and their equivalents set real constraints on what you can send and what you must include.

4. Project Management

Why the Notes app breaks

It’s yours alone. Nobody else can see it, nothing has a due date that reminds anybody, and the client has no visibility — so they ask. And you spend your time answering instead of doing.

What Trello buys you — and where it stops

Shared visibility, free, and honestly fine for a good while.

Where it stops: the free plan caps at 10 boards per workspace. Agencies naturally run a board per client, so you hit a wall at roughly ten clients. Standard is $5/user/mo annual. Calendar, Timeline, Table and Dashboard views are Premium-only at $10/user/mo — so client-facing timelines cost double what you budgeted.

The pick: ClickUp

• Free Forever — $0
• Unlimited — $7/user/mo annual ($10 monthly)
• Business — $12/user/mo annual ($19 monthly)

So if clients are going into the workspace, real client collaboration starts at $7/user/mo. Budget for that from the start rather than discovering it after you’ve invited someone.

5. Invoicing & Getting Paid

Why Gmail breaks

There’s no record of what’s outstanding. You genuinely cannot answer “who owes me money right now” without digging through your sent folder. Nothing chases automatically — so chasing stays a thing you have to decide to do, it feels awkward, you put it off, and forty-five days later you’re annoyed about money you earned in month one.

What Wave buys you — and where it stops

Real invoices, free. Invoicing on Wave is still $0 in 2026. More importantly, the follow-up stops being a personal favour and becomes a system.

Where it stops: automatic bank import, branded invoices, automated payment reminders, and recurring billing are all Pro at $19/mo. The automated reminders are usually the thing you actually wanted.

Fees: 2.9% + $0.60 on cards, 1% on ACH with a $1 minimum and no cap.

The pick: Stripe

No monthly fee. Cards are 2.9% + 30¢. Stripe Invoicing adds 0.4% per paid invoice, capped at $2.00.

Worth noting the comparison: Wave’s ACH is 1% uncapped, so a $10,000 retainer costs $100 there versus $5 on Stripe. On small invoices the difference is noise. On retainers, the caps are the whole game.

6. Client Results & Reporting

Why Google Sheets breaks

You are the report. Every month you pull numbers by hand, and the client receives a grid they can’t interpret — so they ask “is that good?” The value you delivered ends up depending on you being available to explain it.

What Google Slides buys you — and where it stops

Narrative. A report that says what happened and why it matters is worth substantially more than the same numbers in a table, and clients renew on the story, not the spreadsheet.

Where it stops: it’s still entirely manual, every month, every client. At three clients that’s an afternoon. At ten it’s most of a week, forever.

The pick: AgencyAnalytics for most agencies

Three tools do this well. What separates them is the real cost at ten clients, so here are the numbers.

Databox — Free (3 sources, 1 dashboard, 1 user). Agency Starter $79/mo, Agency Pro $159/mo, billed annually.

The catch: Pro includes only 3 data sources. Each additional source is $5.60/mo. Ten clients at three sources each means 27 extra sources — about $151/mo on top. Real cost lands near $310/mo, not $159.

AgencyAnalytics — $20/client/mo annual ($25 monthly). One plan, no tiers. White-label (logo, colours, custom domain) included at base price, unlimited staff and client users. Ten clients ≈ $200/mo all-in. Rank tracking is a paid add-on.

Looker Studio — free, unlimited dashboards and users. Genuinely free. No native white-label and no multi-client automation, so you rebuild per client by hand. Free in licence, expensive in labour.

7. Hiring

Why Fiverr breaks

You’re buying a deliverable, not a person. Fine for a logo. Useless for anything needing context about your business, because every task restarts from zero.

The fee structure also punishes the exact pattern agencies use: 5.5% plus a flat $3.50 on orders under $200. A $30 task costs $35.15 — an effective rate of about 17%. If you’re using Fiverr, batch small jobs into one larger order.

What Upwork buys you — and where it stops

An actual working relationship with someone who learns your business. Rates run roughly $10–20/hr for VAs, $15–35 for designers and SEO, and $15–50 for developers, per Upwork’s own published data.

Fees: 5% marketplace fee (3% for eligible US clients paying by bank), plus a Contract Initiation Fee of $0.99–$14.99 per contract — charged even with freelancers you’ve already hired.

Where it stops: hiring is slow. Post, screen, interview, trial, onboard — four to six weeks before someone is genuinely useful. And once they’re in, the work still needs managing, which is your time.

The pick: Claude

$20/mo. Not instead of people — for the layer of work that’s high-volume, rule-based, and doesn’t need human judgment. That’s the first-draft layer, and it’s where a surprising share of junior hours actually go.

The highest-leverage use isn’t asking for output. It’s building the brief before you hire:

I’m about to hire a [role] for my [type] agency. Before I write the job post, interrogate me: ask me the questions a good operator would ask to work out what this role actually needs to own, what it should never touch, and what “good” looks like at 30, 60 and 90 days. Then write the job post, a screening task that reveals skill rather than effort, and 8 interview questions that would catch someone who interviews well but can’t do the work.

Most bad hires are bad briefs. This fixes the brief.

What it actually costs

Solo operator, 0–3 clients

Small agency, ~10 clients, 3 people

Notice that reporting is over half the bill. It’s the single most expensive system in the stack, and the one where the tool choice moves the number most. Worth deciding deliberately rather than defaulting.

The order to fix them

You cannot rebuild seven systems at once, and trying is how people end up with four half-migrated tools and no working process. Order by what’s costing you money today:

1. Invoicing. The only one that directly changes your cash position. Moving retainers to ACH and switching on automated reminders can pay for everything else on this list.
2. Contracts. Cheap to fix, and the only one where the downside is unbounded. A single bad scope clause costs more than every subscription here combined.
3. Onboarding. Best ratio of hours saved to effort spent. One good intake form kills the back-and-forth week.
4. Project management — but only once more than one person touches a client. Before that it’s overhead pretending to be progress.
5. Reporting. Expensive in both money and setup time. Worth it around five clients, painful before.
6. Outreach. Not because it doesn’t matter — because a lead-gen system pointed at broken delivery just fills a leaky bucket faster.
7. Hiring. Last, deliberately. Hire into systems that already work. Hiring into chaos only makes the chaos more expensive.

The expensive traps, in one place

• Typeform Basic caps at 100 responses/mo, same as free. The $29 buys payments, not volume.
• Mailchimp’s AUP bans cold outreach outright, including scraped or enriched lists. Termination, not a warning.
• Sales Navigator doesn’t raise connection limits — only search and InMail.
• Trello free stops at 10 boards, which for most agencies means 10 clients.
• ClickUp free guests get full edit rights. Never put a client in a free workspace.
• Stripe retainers stack Payments + Billing to ~3.6%. ACH caps it at $5.
• Databox Pro includes only 3 data sources; the rest are $5.60/mo each.
• Upwork charges a contract initiation fee per contract, even for freelancers you’ve hired before.
• Fiverr adds a flat $3.50 under $200 — about 17% on a $30 task.
• Docusign Standard allows ~8 envelopes per user per month, despite the annual framing.

What none of this solves

Everything above reduces friction. None of it does the work.

Apollo builds the list — someone still writes the follow-ups. ClickUp holds the tasks — someone still moves them. AgencyAnalytics pulls the numbers — someone still works out what they mean and tells the client in a way that makes them renew. Better tools take the admin out. The judgment work stays exactly where it was, which is on you.

That’s the ceiling on tooling, and it’s roughly where most agencies plateau: the owner becomes the bottleneck for everything that requires actual thinking, and no subscription fixes that.

The step past better tools is work that runs without you in it — trained on your business, your voice, your process, handling the layer you’d otherwise hire a junior for.

Prices verified August 2026. Tool pricing changes constantly — check the current page before you commit to an annual plan.